JP2015506621A - Display processor for 3D display - Google Patents

Abstract

Obtaining an original image from the 3D image data, deriving a derived image from the original image, generating a first series of images including the derived image, (i) obtaining a further original image from the 3D image data; Generating a second series of images comprising the original image, or (ii) obtaining further original images from the 3D image data, deriving further derived images from the further original images, further derived images and further A second series of images including at least one of the original images is generated, and the number of first-mentioned derived images is greater than the number of further derived images compared to the total number of each series of images, and the stereoscopic view 110 is generated. A display processor (120) that provides a first series of images to the 3D display 140 to provide a second series of images to the 3D display to provide a pseudoscopic field of view 112.

Description

  The present invention relates to a display processor for display on a 3D display and a method for processing three-dimensional [3D] image data. The invention further relates to a 3D display comprising said display processor, a tablet device, a digital photo frame and a smartphone and a computer program product for carrying out said method.

  3D displays, especially television receivers equipped with 3D displays, are becoming more popular among consumers as 3D displays provide stereoscopic depth perception to viewers. So-called autostereoscopic 3D displays provide the stereoscopic depth perception without requiring the viewer to wear deflected or shutter-based glasses. To that end, an optical component such as a lenticular lens array is used, whereby the display can emit a viewing angle comprising at least a left view and a right view of the scene from each predetermined point on the 3D display. This allows the observer to see different images for each eye when positioned accordingly within the viewing angle.

  Some autostereoscopic displays, sometimes referred to as automatic multi-view displays, provide multiple views of the same scene, not just left and right views. This further provides a three-dimensional recognition of the scene while allowing the viewer to assume multiple positions within the viewing angle, i.e. move left and right in front of the display.

  However, not all positions envisioned by the observer are equally suitable for obtaining a three-dimensional recognition of the scene. In particular, if the display is arranged to repeat the viewing angle as a series of viewing angles, for example, the left eye perceives the outermost right field of view of a given viewing angle and the right eye is the outermost of the adjacent viewing angles The viewer may be positioned to perceive the left field of view. At such a viewing position, the observer gets a more false so-called pseudoscopic perception of the scene, often where the scene appears to have depth. Pseudoscopic perception is known to cause headaches and other signs of visual distortion.

  Furthermore, at the aforementioned viewing position, the difference between both fields of view is greater than that of the other viewing positions at a series of viewing angles. As a result, any visual distortion is further exacerbated. The observation position will hereinafter be referred to as a super pseudoscopic observation position. The super pseudoscopic viewing position not only provides a pseudoscopic field of view, but also provides a difference corresponding to the two most different fields of view of two adjacent viewing angles (eg, those at the left and right ends). The viewing position is also sometimes referred to as a super pseudoscopic viewing position.

  WO 2005/091050 describes a multi-view display device for displaying a plurality of fields of view, with a plurality of fields of view having respective viewing angles with respect to the displayed object. The display device comprises optical means for displaying a plurality of viewing angles and drive means for providing the optical means with a set of image data corresponding to each field of view, wherein the first of the plurality of viewing angles is provided. One has an angular distribution of the field of view for the display device.

  The set of image data has an angular distribution of a first portion of the field of view adjacent to the increasing viewing angle and a second portion of the field of view adjacent to the decreasing viewing angle, and a maximum field corresponding to the maximum viewing angle; It is provided that the angular distribution has an initial one of the fields of view between the minimum field of view corresponding to the minimum field of view.

  As a result, the multi-field display device described above includes a first portion of the adjacent field of view that provides stereoscopic viewing to the viewer, and a second portion of the adjacent field of view that provides the viewer with a pseudoscopic field of view. I will provide a. It is said that super pseudoscopic regions can be prevented by creating several pseudoscopic images.

International Publication No. 2005/091050 US Pat. No. 6,064,424 International Publication No. 2006/117707 International Publication No. 2007/063477

  The problem with the multi-view display device described above is that the quality of the field of view of the multi-view display device provided to the observer is insufficient.

  It would be advantageous to have a method that allows improved display of 3D image data on a display processor or 3D display.

  In order to better solve this problem, the first aspect of the present invention is on a 3D display arranged to emit a series of fields of view of 3D image data adjacent to each other in a series of repeated viewing angles. Display processor for processing 3D [3D] image data for display at a display processor that allows stereoscopic viewing of 3D image data at a plurality of viewing positions within a viewing angle of a series of fields of view. A first series for a 3D display to be emitted as a first part of a series of fields of view to provide said stereoscopic view of 3D image data at a plurality of viewing positions within each viewing angle. A series of fields of view that provide at least a pseudoscopic field of view of 3D image data at additional viewing positions within each field of view. Providing a second series of images to the 3D display to be emitted as a second part of the series of fields of view adjacent to the first part of the image, obtaining an original image from the 3D image data, Deriving derived images, generating a first series of images comprising the derived images, (i) obtaining a further original image from the 3D image data and generating a second series of images comprising the further original image Or (ii) obtaining a further original image from the 3D image data, deriving a further derived image from the further original image, and comprising a second series of at least one of the further derived image and the further original image An image is generated and a display processor is provided that is arranged to have the number of derived images initially described be greater than the number of further derived images compared to the total number of each series of images.

  In a further aspect of the invention, a 3D display is provided with a specified display processor. In a further aspect of the invention, a tablet device, digital photo frame, or smartphone is provided with a specified mobile display device.

  In one aspect of the invention, a method of processing three-dimensional [3D] image data for display on a 3D display arranged to emit adjacent to each of a series of repeating viewing angles. In each of the viewing angles of a series of repetitive fields, a method is provided in which a series of fields of 3D image data enables stereoscopic viewing of 3D image data at a plurality of viewing positions within each field of view. The method includes a first series of images for a 3D display to be emitted as a first part of the series of views to provide the stereoscopic view of 3D image data at a plurality of viewing positions within each viewing angle. And a second of a series of fields adjacent to a first portion in the series of fields that provide at least a pseudoscopic field of 3D image data at a further viewing position within each field angle. As a part, providing a second series of images to a 3D display to be emitted, obtaining an original image from the 3D image data, deriving a derived image from the original image, and comprising a first image comprising the derived image Generating a series of images; (i) obtaining a further original image from the 3D image data and generating a second series of images comprising the further original image; or (ii) Obtaining a further original image from the D image data, deriving a further derived image from the further original image, and generating a second series of images comprising at least one of the further derived image and the further original image; The first mentioned number of derived images comprises the step of making the number of further derived images larger than the total number of each series of images.

  In a further aspect of the invention, a computer program product is provided with instructions for causing a processor system to perform a specified method.

  The foregoing means provides a display processor for processing 3D image data for display on a 3D display. The 3D image data is image data that provides stereoscopic viewing. That is, it enables each of the observer's eyes to perceive a slightly different field of view of the scene included in the image data. As a result, the viewer is provided with a sense of depth. A 3D display is a so-called autostereoscopic multi-view display that typically comprises optical means for emitting from any point on the 3D display adjacent to a series of views of 3D image data. A series of fields of view is emitted in the form of a field of view generated from a 3D display, the field of view is one of a series of repeated field angles, and each field of view of the series of repeated field angles is a series of the aforementioned Emitted by a 3D display with a field of view and with different angular orientations relative to other viewing angles.

  The display processor provides a first portion of the field of view that provides a stereoscopic view of the 3D image data at each of a plurality of viewing positions of the viewing angle as part of the series of views, thereby providing a stereoscopic view at each of the viewing angles. Arranged to provide an observation area. Thus, at multiple viewing positions within each of the viewing angles, the viewer perceives a slightly different field of view of the scene in the 3D image data with each of his eyes due to differences in the field of view providing a sense of depth. be able to. Conceptually, the field of view in the first part corresponds to the field of view acquired by the camera facing the scene contained in the 3D image data and moving from left to right in front of the scene and relative to the scene. Create a series of fields of view.

  In addition, the display processor provides a second portion of the field of view that provides a pseudoscopic field of view of the 3D image data at at least further viewing positions at each of the viewing angles as part of the series of fields of view. Arranged to provide a pseudoscopic viewing area at each of the corners. A pseudoscopic field of view, known as pseudo-stereoscopic vision, refers to each of the observer's eyes perceiving a slightly different field of view of the scene in the 3D image data, but the field of view is reversed with respect to the stereoscopic vision. For example, a pseudoscopic field is acquired when a field of view intended to be viewed by the right eye is viewed by the left eye, and a field of view intended to be viewed by the left eye is viewed by the right eye. As a result, the viewer is provided with an inverted depth sensation and thus typically an unnatural depth sensation.

  The second part is adjacent to the first part in the series of fields of view. As a result, the observer can seamlessly transition from stereoscopic viewing to a pseudoscopic visual field by moving from the stereoscopic viewing region to the pseudoscopic viewing region. Inevitably, the pseudoscopic viewing area is located over the second portion of the field of view, for example within the adjacent viewing angle, in the stereoscopic viewing area adjacent to the rightmost field of view in the predetermined stereoscopic viewing area. As the difference or difference between the leftmost field of view is dispersed, the super pseudoscopic viewing area is reduced or avoided. Thus, by deliberately providing a pseudoscopic viewing area adjacent to the stereoscopic viewing area, other perceived differences within the super pseudoscopic viewing position are distributed across many fields of view. Note that the difference, and thus the associated visual distortion, decreases with an increasing field of view in the second part.

  The display processor is further arranged to generate a first series of images to be emitted as the first part of the series of fields of view. For this purpose, the display processor is arranged to directly acquire the original image from the 3D image data. Acquiring the image may comprise, for example, performing a rendering of the field of view of the 3D image data when the 3D image data is provided in a so-called image + depth format, or acquiring the image from a field of view renderer. . Acquiring the image also comprises performing a synthesis of the field of view of the 3D image data, for example, when the 3D image data is provided in a so-called left + right image format, or acquiring the image from a field of view synthesizer. May be. The image may also be obtained directly from 3D image data when the image data is provided in a multi-view format (ie comprising a series of views). The term original image refers to each image showing a different field of view of the scene (ie, corresponding to a different position of the camera relative to the scene) contained within the 3D image data, and each different field of view of the scene is derived directly from the 3D image data. Note that you do.

  The display processor is further arranged to derive a derived image from the original image. Here, deriving and deriving are, for example, generating interpolated images between original images and / or generating extrapolated images next to the original images, and / or Refers to extrapolation. The display processor further generates a first series of images to include at least the derived images. Note that the first series of images may or may not include one or more original images in addition to the derived images. Unlike the interpolated image and / or extrapolated image, the original image is acquired directly from the 3D image data, while the latter is derived from the original image and is not acquired directly from the 3D image data.

  The display processor obtains the additional original image from the 3D image data to generate a second series of images including the additional original image or updates from the 3D image data in other similar manners as described above. Is further arranged to derive a second series of images including obtaining the original image, deriving the further derivative image from the further original image, and including at least one of the further derivative image and the further original image . Again, deriving and deriving are interpolating, for example generating interpolated images between further original images and / or generating extrapolated images next to further original images And / or extrapolation. Further, if the second series of images includes further derived images, the number of initially derived images is more than the total number of each image in each series of images than the number of further derived images. large.

  According to the foregoing means, the first part of the series of fields of view further comprises more interpolated and / or extrapolated images than the second part of the series of fields compared to the total number of images in each part. There is an effect that it is composed. Thus, the field in the first part is more likely to be an interpolated field and / or an extrapolated field than the second part. Accordingly, the observer will encounter a relatively greater number of interpolated and / or extrapolated images in the stereoscopic viewing area than the respective pseudoscopic viewing areas of the viewing angle.

  The present invention is based in part on the realization that two otherwise unrelated techniques can be advantageously combined. The acquisition of the original image using, for example, field rendering or field synthesis is typically more computationally complex than the acquisition of images interpolated and / or extrapolated from the original image, so the first This technique involves the use of interpolation and / or extrapolation to reduce the computer load in generating images for 3D displays. The second technique relates to the introduction of a pseudoscopic observation region in order to avoid the super pseudoscopic observation region or to reduce the size of the super pseudoscopic observation region. The inventor recognizes that it is desirable to obtain a smooth transition between pseudoscopic fields within the second part of the series of fields so that the visual distortion caused by the pseudoscopic field itself is not further increased. did. If the field of view includes interpolated and / or extrapolated images, this may interfere with smooth transitions between the fields of view. Thus, when reducing the computational complexity of generating an image to be emitted as a series of fields of view by using interpolation and / or extrapolation to generate an interpolated image and / or extrapolated image, the second series It is more desirable to apply the interpolation and / or extrapolation when generating the first series of images than when generating the first image.

  Optionally, the display processor is arranged to derive a derived image from the original image using zero order interpolation techniques and / or zero order extrapolation techniques. The term zero order refers to repeating the original image to produce an interpolated and / or extrapolated image. Image repetition involves less computational complexity. Since the repetition of the original image avoids the artificial effects of interpolation and / or extrapolation, the sharpness of the original image is maintained in the interpolated and / or extrapolated image. Advantageously, if the 3D display exhibits optical crosstalk between adjacent fields of view, for example, the 3D display is a so-called fractional view display where the field of view essentially has significant crosstalk due to adjacent fields of view. In this case, the sharpness of the image within the predetermined field of view is maintained when the adjacent field of view comprises an interpolated image and / or extrapolated image that is a repeat of the image within the predetermined field of view.

  If necessary, the display processor is arranged to derive further derived images from the further original images using first order or higher order interpolation and / or extrapolation techniques. Interpolated images and / or extrapolated images created using first-order or higher-order interpolation and / or extrapolation techniques are typically interpolated images and / or extrapolated created using zero-order interpolation. Provides a transition between the original images that is smoother than the images. Advantageously, if the second series of images includes an interpolated image and / or an extrapolated image, the smooth transition between the pseudoscopic fields in the second part of the series of fields can be interpolated and / or Maintained despite the use of extrapolation.

  Optionally, the display processor is arranged to acquire further original images by selecting all or a subset of the original images. The computational complexity for obtaining additional original images can be reduced by reusing all or a subset of the original images from the first series of images in the second series of images. Advantageously, the computational complexity of performing field rendering or field synthesis is reduced if at least some of the original images are reused in the second series of images.

  Optionally, the second series of images is composed of all or a subset of the original image. Therefore, since all further original images are acquired from the original image, it is not necessary to acquire any original images separately from the 3D image data. Advantageously, it is not necessary to perform any visual field rendering or visual field synthesis to obtain additional original images, especially from 3D image data.

  If necessary, the display processor is arranged to blur the second series of images. By blurring the second series of images, a sharper transition between the pseudoscopic fields in the second part of the series of fields is obtained because the sharpness of the images is reduced, so that Make the transitions inconspicuous. Advantageously, the artificial effects of interpolation and / or extrapolation in the interpolated image and / or extrapolated image are reduced by blurring.

  Optionally, the second series of images can be displayed by applying a spatial low pass filter to the display processor, an individual one of the second series of images, or by averaging multiples of the second series of images. It may be arranged to blur the image. The spatial low pass filter blurs the image separately. That is, other image pixels are not considered. In contrast, averaging of multiple images blurs the image by averaging pixels across multiple images. Both techniques are suitable for blurring the second series of images.

  Optionally, the first adjacent subset of the series of fields and the second part of the series of fields together form a series of fields. The pseudoscopic viewing area provided within each of the repeated viewing angles thus always forms a transition between the stereoscopic viewing areas of adjacent viewing angles.

  Optionally, the 3D display is arranged to emit a series of fields of view as a series of fractional views, each having a series of fractional views that exhibit optical crosstalk with O adjacent fractional views. Therefore, the 3D display is a so-called fractional view display. Such a display is typically referred to as a P / Q display, where P indicates the number of fractional views provided for each viewing angle in a series of repeating viewing angles, and Q is optical Indicates the number of fractional views visible to the user when viewing a single one of the fractional views due to crosstalk. For example, if the fractional view display is a 20/3 display, Q is equal to O + 1, and the viewer will perceive two adjacent fractional views as well when viewing a given fractional view, As a result, the observer perceives a total of three fractional views. Note that O = 2, Q = 3.

  Optionally, the display processor (i) derives the first series of images by deriving O derived images for each of the original images using zero order interpolation techniques and / or zero order extrapolation techniques. And (ii) arranged to generate a second series of images that do not contain further derived images. Thus, for each original image, O interpolated and / or extrapolated images are generated by iteration, where O is an adjacent fractional view that exhibits optical crosstalk by one in a series of fractional views. Is equal to the number of As a result, an observer may be under the influence of optical crosstalk of at least one adjacent fractional view with repetition of the original image when perceiving one original image of the fractional view. Therefore, the observer does not perceive optical crosstalk very much. Optical crosstalk may also be O adjacent fractional views, each containing the repeated image. Thus, the observer will not perceive any crosstalk or will perceive a trivial amount of crosstalk. The use of zero-order techniques therefore results in optical crosstalk that is not clearly visible, and a first series of fractional view images that appear sharper on average. Therefore, the observer obtains an improved image quality within the stereoscopic observation area. Furthermore, the second series of images does not include interpolated images and / or extrapolated images. Therefore, a smooth transition between the fractional views in the pseudoscopic observation area is obtained. Furthermore, as adjacent fractional views are mixed together, optical crosstalk between adjacent fractional views further increases the smoothing.

  Optionally, the first series of images includes substantially O + 1 times the total number of the second series of images. For example, for a 3D display in which each of the fields of view exhibits optical crosstalk with two adjacent fields of view (ie, O = 2), the first series of images is the same number of 3 ( That is, 2 + 1) times as many images are included. As a result, the first part of the series of fields of view is three times larger than the second part of the series of fields of view one by one of the series of repeated viewing angles. Advantageously, the transition between adjacent but different views is substantially equal in each of the first and second portions of the view. Advantageously, the same amount of depth perception may be obtained in each of said portions of the field of view, for example instead of a second portion of the field of view providing greater depth perception than the first portion of the field of view.

  Therefore, the above means can be used for the stereoscopic observation region by repeating the original image, particularly by repeating the original image in the same number as the number of adjacent fractional views that exhibit optical crosstalk in each of the series of fractional views. This has the effect of reducing optical crosstalk between adjacent fractional views. Furthermore, no repetition is utilized within the pseudoscopic viewing area (ie, optical crosstalk is not reduced) to increase the smoothness of transitions between fractional views within the pseudoscopic viewing area.

  Optionally, the first part of the series of fields of view is at least twice as large as the second part of the series of fields of view. Therefore, a stereoscopic observation area that is at least twice as large as the pseudoscopic observation area is provided.

  It will be appreciated by those skilled in the art that two or more of the above embodiments, embodiments and / or aspects of the invention may be combined in any way deemed useful. Modifications and variations of the 3D display, tablet device, digital photo frame, smartphone, method, and / or computer program product corresponding to the described display processor variations and variations may be implemented by one of ordinary skill in the art based on this description. it can. The invention is defined in the independent claims. Useful options are defined in the dependent claims.

  These and other aspects of the invention are apparent from and will be elucidated with reference to the embodiments described hereinafter.

Fig. 4 shows a display processor according to the invention and a 3D display for emitting a series of fields of view adjacent to each other in a series of viewing angles. Fig. 4 shows a schematic view of the viewpoint provided by each of a series of fields of view as a function of observation positions at a series of viewing angles. Fig. 3 shows a schematic view of the viewpoint provided by a series of fields of view with a single viewing angle, the series of fields of view providing a stereoscopic view. FIG. 4 shows a schematic view of a series of fields of view including a first part for stereoscopic viewing and a second part for pseudoscopic field of view. Fig. 2 shows a schematic view of a series of fields of view comprising a first part and a second part according to the invention. Fig. 4 shows a further example of a series of fields of view comprising a first part and a second part according to the invention. Fig. 4 shows a further example of a series of fields of view comprising a first part and a second part according to the invention. 1 shows a tablet device comprising a display processor according to the invention and a 3D display. 1 shows a method according to the invention. 1 illustrates a computer readable medium comprising a computer program product according to the present invention.

  FIG. 1 shows a display processor 120 connected to a 3D display for providing a series of images 122 to the 3D display 140. The 3D display 140 is an autostereoscopic 3D display that enables stereoscopic viewing of content displayed on the 3D display 140 without the need for the user to wear glasses. The 3D display 140 includes a light generating portion 142 that is typically composed of an array of light emitting elements or light modulating elements. For example, the light generating portion 142 may be formed by a liquid crystal display (LCD) panel and a backlight, as is well known from the display art.

  The 3D display 140 further comprises optical means 144 for redirecting the light generated by the light generating portion 142 in a different direction. The light generating portion 142 may be suitably arranged and cooperate with the optical means 144 such that a series of fields 0-5 are emitted from the 3D display 140 in the form of a viewing angle 104. Further, the 3D display 140 may be arranged to emit adjacent images within a series of fields 0-5 when a series of images 122 are provided. Thus, an observer will perceive one of each of the series of images 122 when viewing one of the series of fields of view 0-5. The series of images 122 may correspond to a camera that faces a scene included in the 3D image data and moves from left to right in front of the scene and relative to the scene. Thus, an observer 110 located within the viewing angle 104 and perceiving two different things 0, 1 in a series of views 0-5 can obtain a stereoscopic view of the scene.

  Note that the above configuration of the 3D display and the method of processing a series of images 122 for display as a series of views 0-5 are known per se. For example, US Pat. No. 6,064,424 discloses an autostereoscopic display device having a lens-shaped element as the optical means 144, and a display element (ie, a light emitting element or a light modulating element) and a lens-shaped element. Explains the relationship between. Autostereoscopic displays are also known to include a parallax barrier as the optical means 144.

  FIG. 1 shows a viewing angle 104 that is one of the center of a series of repetitive viewing angles 100, where each of the viewing angles 102, 104, 106 comprises a series of views 0-5. It should be noted that the repeated viewing angle 104 can be a desired as well as an intrinsic property of the optical means 144 of the 3D display 140. Field angle repetition is described in more detail in the aforementioned US Pat. No. 6,064,424.

  The observer is shown in FIG. 1 at two observation positions. At the first observation position 110, the observer perceives the first visual field 0 with his / her left eye while perceiving the second visual field 1 with his / her right eye. The aforementioned coincidence of the series of images 122 to the camera, moving in front of the scene and from left to right with respect to the scene, gives the observer a stereoscopic view at the first viewing position 110. Accordingly, the first observation position 110 is the stereoscopic observation position 110. At the second observation position 112, the observer perceives the third visual field 5 with his / her left eye while perceiving the fourth visual field 0 with his / her right eye. The observer will thus obtain a pseudoscopic field of view at the second viewing position 112, ie, the viewing position is the pseudoscopic viewing position 112. In this case, the third visual field 5 corresponds to the rightmost visual field of the central viewing angle 104, and the fourth visual field 0 corresponds to the leftmost visual field of the right viewing angle 106. As a result, a very large and inverted depth sensation is obtained at the second viewing position, i.e. the viewing position is the super pseudoscopic viewing position 112.

  FIG. 2 shows a schematic diagram of a series of fields across each of a series of repeated viewing angles. The horizontal axis is a series of values for each repeated viewing angle 102, 104, 106 in an order corresponding to an observer moving from left to right in front of the 3D display 140 parallel to the display surface of the 3D display 140. Each of the fields of view 0-5 is shown. That is, the observer traverses through a series of fields of left viewing angle 102, central viewing angle 104, and finally right viewing angle 106. The vertical axis corresponds to a viewpoint 160 obtained by an observer who perceives one of a series of fields of view 0-5, where the viewpoint 160 is for a scene included in the 3D image data. Here, a low value (ie, a low position on the vertical axis) corresponds to the left-hand viewpoint for the scene, and a high value (ie, a high position on the vertical axis) corresponds to the right-hand viewpoint for the scene. FIG. 2 thus shows a viewpoint 160 that changes from left to right with respect to the scene as the observer traverses through a series of views 0-5 within the left viewing angle 102, which is then centered. When traversing through a series of views 0-5, such as viewing angle 104, it jumps again to the left and changes to the right. In addition, the aforementioned stereoscopic viewing position 110 and super pseudoscopic viewing position 112 are shown. It is clear from FIG. 2 that the observer obtains a very large and inverted depth sensation at the super pseudoscopic observation position 112 due to a large difference in the viewpoint.

  FIG. 3 shows a schematic diagram of another series of fields of view 0-19. Here, the series of views 0-19 are represented solely for a single viewing angle 108, ie, for reasons of clarity, no repeated viewing angles are shown. In contrast to the series of fields 0-5 shown in FIGS. 1 and 2, the series of fields represented in FIG. 3 is composed of 20 fields. The 3D display 140 may be arranged to emit a series of views 0-19 as a series of fractional views. Here, the term fractional indicates that each one of a series of fields 0-19 exhibits optical crosstalk with adjacent fields. That is, the observer essentially perceives the overlap of a series of fields of view 0-19. Therefore, the 3D display 140 may be a so-called fractional view type 3D display. Note that fractional view 3D displays and methods for processing a series of images 122 for display as a series of views 0-19 are known per se. For example, WO 2006/117707 discloses a stereoscopic display device having a group of lenses as an optical directory means with lens tilt and pitch selected to provide the fractional view. The 3D display 140 may be a so-called 20/3 display, where “20” indicates the number of fractional views emitted within each viewing angle and “3” indicates the degree of crosstalk between the fractional views. Here, the number “3” should be understood to refer to an observer who perceives a total of three of the fractional views when viewing one of the fractional views. Thus, optical crosstalk appears to make two additional and adjacent fractional views visible in any fractional view.

  Of course, it is noted that the 3D display 140 may also have other suitable configurations (ie, 5 fields of view, 9 fields of view, 20 fields of view, other numbers of field of view displays). I want. Further, the 3D display 140 may or may not be configured to generate a series of views 0-19 as a series of fractional views. However, in the following, it is assumed that the 3D display 140 is configured as the above-described 20/3 fractional view 3D display.

  FIG. 3 shows a series of views 0-19 corresponding to a monotonically increasing viewpoint 160. That is, the viewpoint 160 increases monotonically between the first field 0 and the last field 19 in the series of fields 0 to 19. As in the case of the series of fields 0-5 shown in FIGS. 1 and 2, the observation position is that the observer obtains a super pseudoscopic field of view (ie, the first field of view 0 and itself with his right eye). Between the viewing angle 108 and the adjacent viewing angle (perceived by the left eye) and the final viewing field 19 (or vice versa). With optical crosstalk, the observer will further perceive the final field of view 19 and the second field of view 1 with his right eye and the neighbor of the first field of view 0 and the final field of view 18 with his left eye. Please note that. Note that optical crosstalk causes blurred perception. As a result, visual distortion caused by a very large and inverted depth sensation at the super pseudoscopic viewing position is reduced. However, significant visual distortion remains due to the large depth perceived by the observer.

  FIG. 4 shows a series of fields 0-19 configured to reduce the visual distortion caused by a super pseudoscopic field between adjacent viewing angles. In this figure, the viewpoint 160 corresponding to each of the series of fields 0 to 19 and represented by the dot height on the vertical axis is further numerically represented by the series of viewpoints 162. In numerical notation, a low or lower number corresponds to a left-hand viewpoint or a further left-hand viewpoint 160 for the scene, and a high or higher value corresponds to a right-hand viewpoint or a further right-hand viewpoint 160 for the scene. Note that the viewpoint 160 provided is shown in FIG. Thus, the numeric notation serves to illustrate the relative differences in the viewpoint 160 between a series of fields of view 0-19, but is not an absolute measure.

  FIG. 4 shows the first parts 0 to 14 in which the stereoscopic view is provided (that is, the viewpoint 160 is changed from the viewpoint “0” in the first field 0 to the viewpoint “0” in the final field 14 of the first parts 0 to 14). A series of fields of view 0-19 are shown, including monotonically increasing to 14 ”. Further, the series of views 0-19 is a second portion 15-19 in which a pseudoscopic view is provided (i.e., the viewpoint 160 is changed from the viewpoint "11" in the first view 15 to the second portions 15-19. Of the final visual field 19 of the final visual field 19 decreases monotonously to the viewpoint “2”). Therefore, an observer who traverses through a series of fields 0 to 19 within the viewing angle 108 notices the stereoscopic viewing region and the pseudoscopic viewing region within the viewing angle 108. Further, the steps in the viewpoint 160 between each of the fields of view in the second portions 15-19 are chosen such that changes in the viewpoint 160 in the first portions 0-14 substantially cancel. That is, an observer traversing through a series of fields of view 0 to 19 obtains substantially the same viewpoint on the right side of the viewing angle 108 as the left side of the viewing angle 108. In this case, the reduction of the viewpoints in the second parts 15-19 results in the viewpoints in the first parts 0-14 as a result of the first parts 0-14 having about 2.5 times as many fields of view. The average is selected to be about 2.5 times larger than the increase.

  Compared to the series of fields 0-19 provided in FIG. 3, the series of fields 0-19 provided in FIG. 4 not only have a small stereoscopic viewing area, but also the leftmost field of the stereoscopic viewing area. Any super pseudoscopic between adjacent viewing angles as a result of the introduction of a pseudoscopic viewing region between stereoscopic viewing regions, providing a step transition to the rightmost visual field of the stereoscopic viewing region within the adjacent viewing angle from Note that the observation position is also avoided. Thus, further viewing positions within the viewing angle 108 provide a pseudoscopic field here, but the very large and inverted depth sensation of FIG. 3 is avoided.

  FIG. 5 shows a series of views 0-19 as provided by the present invention. Again, the series of views 0-19 includes a first portion 0-14 in which a stereoscopic view is provided and a second portion 15-19 in which a pseudoscopic view is provided. The second portions 15-19 are shown as in FIG. 4, and as shown in FIG. 4, only the first field of view 15 shows the viewpoint “10” instead of the viewpoint “11”. However, the first parts 0 to 14 here are repetitive fields in that the fields 0, 1, and 2 indicate the same viewpoint “0”, the fields 3, 4, and 5 indicate the same viewpoint “3”, and the like. including. Here, the term repetition refers to the same viewpoint (ie, a field of view showing the same image of the scene). Obviously, the overall tilt of the viewpoint in the first portions 0-14 is substantially similar to the tilt of FIG. That is, however, the first portions 0-14 provide a substantially substantially similar transition within the viewpoint 160, albeit with a coarser step size. As a result of the 3D display 140, which is a 20/3 fractional view 3D display in this example, the viewer is, on average, less affected by optical crosstalk. For example, when perceiving the field 4 of the series of fields 0 to 19 represented in FIG. 4, the observer will further perceive the fields 3 and 5. Therefore, the user perceives a mixture of the viewpoint “3”, the viewpoint “4”, and the viewpoint “5”. In contrast, when perceiving the field 4 of the series of fields 0-19 represented in FIG. 5, the observer will perceive the same viewpoint “3” repetition further. Thus, the observer will not perceive or noticeably perceive a mixture of different viewpoints. A mixture of different viewpoints typically results in a blurry impression (ie loss of definition). By avoiding mixing of different viewpoints, the loss of definition is eliminated or reduced. Note that a field of view provides a mixture of different viewpoints (eg, in field of view 5 the viewer will perceive a mixture of viewpoints “3”, “3” and “6”). . However, on average, the first portions 0-14 of FIG. 5 provide less blur (ie, more sharpness) than the first portions 0-14 of FIG. At the same time, the second portions 15-19 of FIG. 5 do not have the same amount of blurring as the second portions 15-19 of FIG. Provide similar sharpness loss).

  5 may be obtained by the display processor 120 providing a first series of images to the 3D display 140 to be emitted as the first portion 0-14 of the series of views 0-19. Good. Display processor 120 generates a first series of images comprising X and X (X is greater than zero) interpolated images for each of the original image and the original image, and interpolates the original image using a zero order interpolation technique. The original image is acquired from the 3D image data and arranged to interpolate the original image. In the example of FIG. 5, X is equal to 2, ie, for each of the original images corresponding to viewpoints “0”, “3”, “6”, “9” and “12”, two interpolated images are present. Are generated and correspond to a series of viewpoints “0, 3, 6, 9, 12”, respectively, and a series of viewpoints “0, 0, 0, 3, 3, 3, 6, 6, 6, 9, 9, 9”. , 12, 12, 12 ", resulting in a series of original images that are interpolated to provide a first series of images. Thus, the five original images are interpolated to obtain a first series of images comprising 5 original images and 10 interpolated images.

  Acquiring the original image from the 3D image data may include performing field of view rendering to acquire an image corresponding to a series of viewpoints “0, 3, 6, 9, 12” Note that interpolating the original image using an interpolation technique may include simply repeating each of the original images twice.

  The second portion 15-19 of FIG. 5 provides a second series of images to the 3D display 140 to be emitted as the second portion 15-19 of the series of views 0-19. It may be obtained by the processor 120. Further, the display processor 120 obtains additional original images from the 3D image data to generate a second series of images including additional original images and Y-interpolated images for each additional original image, and further It may be arranged to interpolate the original image. In the example shown in FIG. 5, Y is zero, ie no interpolated image is generated for each of the further original images. As a result, the generation of a series of views 0-19 to correspond to the series of viewpoints 164 of FIG. 5 is a total of nine (ie, viewpoints “0”, “3”, “6”, “9”, “9”, “ 12), “10”, “8”, “4”, “2”). As a result, the remaining 11 images in the series of fields 0 to 19 are interpolated images.

  FIG. 6a shows another example of a series of fields 0-19 according to the present invention. Again, the series of views 0-19 includes a first portion 0-14 in which a stereoscopic view is provided and a second portion 15-19 in which a pseudoscopic view is provided. First portions 0-14 are shown to be identical to those of FIG. In this example, the display processor 120 may be arranged to obtain additional original images by selection of all or a subset of the original images. In this example, the second parts 15 to 19 are composed of images corresponding to the viewpoints “0”, “3”, “6”, “9”, “12”, and are thus all acquired from the original image. Also good. As a result, the generation of a series of views 0-19 to correspond to the series of viewpoints 166 of FIG. 6a is a total of five (ie, viewpoints “0”, “3”, “6”, “9”, “9”, “ Acquisition of the original image (corresponding to “12”). As a result, the remaining 15 images in the series of views 0-19 are interpolated images.

  Similarly, the first series of images may be composed of images corresponding to a series of viewpoints “0, 0, 2, 2, 4, 4, 6, 6, 8, 8, 10, 10” The second series of images may correspond to a series of viewpoints “8, 6, 4, 2, 0”. In this example, the 3D display 140 may be a 17/2 fractional view 3D display. Similarly, the first series of images is represented by a series of viewpoints “1, 1, 1, 4, 4, 4, 7, 7, 7, 10, 10, 10, 13, 13, 13, 16, 16, 16 , 19, 19 ”, and the second series of images may correspond to a series of viewpoints“ 17, 15, 13, 11, 9, 7, 5, 3 ”. . In this example, only a subset of the original images of the first series of images (ie, the original image corresponding exclusively to viewpoint “7”) is used. The 3D display 140 in this example may be a 28/3 fractional view 3D display.

  FIG. 6b shows another example of a series of fields 0-19 according to the present invention. First portions 0-14 are shown to be identical to those of FIG. In this example, the display processor 120 may be arranged to interpolate further original images using first order or higher order interpolation techniques. For example, Y may be equal to 5/3, i.e., 5/3 interpolated images are generated for each additional original image. For example, further original images may correspond to viewpoints “10”, “6”, and “2”. Here, the additional original images corresponding to the viewpoint “6” may be obtained from the original images of the first portions 0 to 14, whereas the additional original images corresponding to the viewpoints “10” and “2” are used. The resulting original image may be obtained directly from the 3D image data. The further original image interpolated to provide an intermediate image between the images corresponding to viewpoints “10” and “6”, wherein a first-order or higher-order interpolation technique results in an interpolated viewpoint “8 *” The intermediate image corresponds to the interpolation between the viewpoints. Similarly, first-order or higher-order interpolation techniques are used to provide additional intermediate images between the images corresponding to viewpoints “6” and “2”, resulting in an interpolated viewpoint “4 *”. The further intermediate image corresponds to the interpolation between the viewpoints. Thus, a second composed of further original and interpolated images that together form a series of viewpoints 168 that include viewpoints “10, 8 *, 6, 4 *, 2” in second portions 15-19. The second portions 15 to 19 similar to those shown in FIG. 5 may be acquired.

  Note that the primary interpolation of the additional original image may include performing a weighted averaging of the additional original image. Higher-order interpolation analyzes the trajectory of the object between additional original images and recreates the additional original image as faithfully as possible to recreate the object located along the midpoint in the trajectory. Interpolation may be included. It should be noted that such interpolation techniques are known per se from the field of image processing, in particular from the fields of field interpolation and frame rate conversion.

In general, the ratio of interpolation between X and Y may be, for example, 1: 0. That is, the first series of images may include half of the original image and half of the interpolated image, whereas the second series of fields is composed entirely of additional original images. Also good. Similarly, the ratio can be any ratio of 2: 0, 2: 1, 3: 0, 3: 1, 3: 2, 4: 0, 4: 1, 4: 2, and 4: 3, or Other suitable ratios may be used. Note that the interpolation factor, as well as the resulting ratio, need not be an integer. For example, the second series of images by interpolating the first sequence of images to generate a _^5/3 images for each and further the original image by interpolating the two interpolation images for each original image When ^{generated, 2:} _5/3 a is X: the ratio of Y or about 2: 1.67 X: the ratio of Y is obtained.

  Note that the X images may include interpolated images, extrapolated images, or a combination of interpolated and extrapolated images. Similarly, the Y images may include interpolated images, extrapolated images, or a combination of interpolated and extrapolated images. For example, in an alternative to the embodiment whose results are shown in FIG. 6b, viewpoints “8” and “4” may be obtained directly from 3D image data. A primary or higher order interpolation technique may be used to interpolate the further original image to provide an intermediate image corresponding to the interpolation between the viewpoints, resulting in an interpolated viewpoint “6 *”. Good. Similarly, the primary extrapolation technique or the higher-order extrapolation technique results in images corresponding to viewpoints “8” and “4” resulting in extrapolated viewpoint “10 *” and extrapolated viewpoint “2 *”. May be used to provide an additional image next to the image, the additional image corresponding to any viewpoint extrapolation. Therefore, a combination of a further original image and an interpolated image and an extrapolated image that together form a series of viewpoints including the viewpoints “10, 8, 6 *, 4, 2 *” in the second portions 15 to 19 The second portions 15 to 19 similar to those shown in FIG. 5 may be obtained even though the second portions 15 to 19 are configured.

  The display processor 120 can generate the second series of images, for example, by applying a spatial low pass filter to each of the second series of images, or by averaging multiples of the second series of images. May be arranged to blur. Alternatively or additionally, depth dependent blurring may be applied, as is known from WO 2007/063477.

  FIG. 7 shows a tablet device 180 with a display processor 120. That is, the display processor is an internal component of the tablet device 180. The tablet device 180 further includes a 3D display 140. Display processor 120 is shown connected to a 3D display 180 for providing a series of images 122. Alternatively, the display processor 120 may be provided in a digital photo frame or smartphone. The device may also include a 3D display 140. Alternatively, the display processor 120 may be included in the 3D display 140, which constitutes a separate device or a stand-alone device. Alternatively, the display processor 120 may be included in a set-top box, personal computer, game console, or similar device that can be connected to the 3D display 140.

  FIG. 8 illustrates a method 200 for processing 3D image data for a display on a 3D display, wherein the 3D display adjoins a series of fields of 3D image data at each of a series of repeating viewing angles. And a series of fields of view allow stereoscopic viewing of 3D image data at multiple viewing positions within each viewing angle. The method 200 includes a first series of 3D displays to be emitted as a first part of a series of views to provide the stereoscopic view of 3D image data at a plurality of viewing positions within each viewing angle. It includes a first step 220 entitled “Providing a first series of images (PROVIDING A FIRST SERIES OF IMAGES)” including providing images. The method 200 is first for a 3D display to be emitted as a second part of a series of views to provide a pseudoscopic view of 3D image data at (at least) further viewing positions within each viewing angle. Including a second step 240 entitled “Providing a second series of images”, the second part further comprising providing a second series of images. Adjacent to the first portion in the field of view. The method 200 obtains an original image from 3D image data and generates an original image and a first series of images including X interpolated images (X is greater than zero) for each of the original images. It further includes a third step 260 titled “Generating a First Sequence of FIRST SERIES OF IMAGES” comprising interpolating the images. The method 200 obtains additional original images from the 3D image data and interpolates Y (Y is greater than or equal to zero and Y is less than X) for each of the additional original image and the additional original image. A fourth entitled "Generating the Second Series of Images" including interpolating further original images to generate a second series of images containing the images. The step 280 is further included. It is noted that FIG. 7 is not understood to identify the order in which steps 220, 240, 260, 280 need to be performed. In particular, the third step 260 may be performed before the first step 220, and the fourth step 280 may be performed before the second step 240.

  FIG. 9 shows a computer readable medium 300 containing a computer program product 302 for causing a processor system to perform the method according to the invention. The computer program product 302 may be included on a computer-readable medium as a series of machine-readable physical marks and / or as a series of elements having different electrical (eg, magnetic) or optical properties or values. Good.

  It will also be appreciated that the present invention applies to computer programs (especially computer programs on or in a carrier) configured to carry out the present invention. The program may be in the form of source code, object code, code intermediate source, partially compiled object code or any other form suitable for use in carrying out the method according to the invention. Good. It will also be appreciated that there may be many different structural designs for such programs. For example, program code implementing the functionality of the method or system according to the invention may be subdivided into one or more subroutines. Many different ways of distributing functions to these subroutines will be apparent to those skilled in the art. Subroutines may be stored together in one executable file to form a self-contained program. Such executable files may include computer-executable instructions such as, for example, processor instructions and / or interpreter instructions (eg, Java® interpreter instructions). Alternatively, one or more or all of the subroutines may be stored in at least one external library file and linked statically or dynamically (eg, at runtime) with the main program. The main program includes at least one call to at least one of the subroutines. Subroutines may also include function calls with each other. Embodiments relating to a computer program product include computer-executable instructions corresponding to each processing step of at least one of the methods specified herein. These instructions may be subdivided into subroutines and / or stored in one or more files that may be linked statically or dynamically. Another embodiment relating to a computer program product includes computer-executable instructions corresponding to each means of at least one of the systems and / or products specified herein. These instructions may be subdivided into subroutines and / or stored in one or more files that may be linked statically or dynamically.

  A computer program carrier may be any entity element or device capable of transmitting a program. For example, the carrier may include a storage medium such as a ROM (for example, a CD-ROM or a semiconductor ROM) or a magnetic recording medium (for example, a hard disk). Further, the carrier may be a transmissible carrier, such as an electrical signal or an optical signal, which may be transmitted via an electrical or optical cable or by radio or other means. If the program is embodied with such a signal, the carrier may be constituted by such a cable or other device or means. Alternatively, the carrier may be an integrated circuit in which the program is embedded. The integrated circuit is configured or used to perform the associated method. The integrated circuit may be an application specific IC (ASIC). The program may also be incorporated in the form of firmware (ie, as microcode stored within the ASIC) or may be incorporated separately for use by the ASIC.

  It is noted that the above embodiments have been described without limiting the invention and that many other embodiments can be designed by those skilled in the art without departing from the scope of the appended claims. Should do. In the claims, any reference signs placed between parentheses shall also not be construed as limiting the claim. Use of the verb “comprise” and utilization of the verb do not exclude the presence of elements or steps other than those expressly recited in the claims. The indefinite article “a” or “an” immediately preceding a component does not exclude the presence of a plurality of such components. The present invention may be implemented by hardware comprising several separate components and by a suitably programmed computer. In the device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are mutually recited in individual dependent claims does not indicate that a combination of these measured cannot be used to advantage.

Claims (15)

3 for displaying on a 3D display (140) arranged to emit adjacently a series of fields of view (0-5, 0-19) of 3D image data at each of a series of repeated viewing angles (100). With a display processor (120) for processing dimensional [3D] image data, a series of fields of view of 3D image data at a plurality of viewing positions (110) within each viewing angle (102, 104, 106, 108). A display processor (120) that enables stereoscopic viewing,
As a first part (0-14) of a series of fields of view (0-19) for a 3D display to be emitted to provide said stereoscopic view of 3D image data at a plurality of viewing positions within each viewing angle Providing a first series of images,
A second part of a series of fields (0-19) adjacent to the first part in the series of fields, providing at least a pseudoscopic field of 3D image data in a further viewing position within each field of view (15-19) providing a second series of images to the 3D display to be emitted,
Obtaining an original image from the 3D image data, deriving a derived image from the original image, generating a first series of images including the derived image;
(I) acquiring a further original image from the 3D image data and generating a second series of images comprising the additional original image, or (ii) acquiring a further original image from the 3D image data, Deriving further derived images from the image and generating a second series of images including at least one of the further derived images and the further original image, wherein the number of first derived images is the number of each of the series of images. A display processor arranged to be larger than the number of further derived images compared to the total number.   The display processor (120) of claim 1, wherein the display processor is arranged to derive a derived image from the original image using zero order interpolation techniques and / or zero order extrapolation techniques.   3. A display processor according to claim 1 or 2, wherein the display processor is arranged to derive further derived images from further original images using first order or higher order interpolation and / or extrapolation techniques. (120).   4. The display processor (120) according to any one of claims 1 to 3, wherein the display processor is arranged to acquire further original images by selecting all or a subset of the original images.   The display processor (120) of claim 4, wherein the second series of images is composed of all or a subset of the original images.   The display processor (120) according to any of the preceding claims, wherein the display processor is arranged to blur the second series of images.   The display processor applies a spatial low pass filter to individual images of the second series of images or averages the plurality of images of the second series of images, The display processor (120) of claim 6, wherein the display processor (120) is arranged to blur.   A display processor (120) according to any preceding claim, wherein the first part of the series of fields and the second part of the series of fields together form a series of fields.   A 3D display (140) comprising a display processor (120) according to claim 1.   The 3D display according to claim 9, wherein the 3D display is arranged to emit a series of fields of view as a series of fractional views, each having a series of fractional views that exhibit optical crosstalk with O adjacent fractional views. Display (140).   A display processor (120) generates a first series of images by deriving O derived images for each of the original images using (i) zero order interpolation techniques and / or zero order extrapolation techniques. 11. The 3D display (140) of claim 10, arranged to generate a second series of images that do not include any further derived images.   12. A 3D display (140) according to claim 10 or 11, wherein the first series of images comprises substantially O + 1 times the total number of images of the second series of images.   A tablet device (180), a digital photo frame, or a smartphone comprising the display processor (120) according to any one of claims 1-8. A method (200) for processing 3D [3D] image data for display on a 3D display arranged to emit a series of views of adjacent 3D image data at each of a series of repeated viewing angles. A method (200) for enabling stereoscopic viewing of 3D image data at a plurality of viewing positions within a viewing angle of a series of fields of view,
Providing a first series of images to a 3D display to be emitted as a first part of the series of fields of view to provide the stereoscopic view of 3D image data at a plurality of viewing positions within each viewing angle. Step (220);
To be emitted as a second part of a series of fields of view adjacent to a first part of the series of fields of view, providing at least a pseudoscopic field of view of 3D image data at a further viewing position within each field of view Providing (240) a second series of images to the 3D display;
Obtaining an original image from the 3D image data, deriving a derived image from the original image, and generating a first series of images including the derived image (260);
(I) acquiring a further original image from the 3D image data and generating a second series of images including the additional original image (280), or (ii) acquiring a further original image from the 3D image data Deriving further derived images from the further original images and generating a second series of images including at least one of the further derived images and the further original images, wherein the number of first derived images is And a step of making the number of further derived images greater than the total number of images in the series.   A computer program product (302) comprising instructions for causing a processor system to perform the method of claim 14.

Images